# US 12,112,279 B2 · Behavior prediction device

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,112,279 B2
- Document ID: US12112279B2
- Status: Granted
- Kind: B2
- Application number: 17/485,893
- Filed: 2021-09-27
- Published: 2022-06-02
- Granted: 2024-10-08
- Earliest claimed: 2020-11-30
- Assignees: Toyota Jidosha Kabushiki Kaisha
- Applicants: Toyota Jidosha Kabushiki Kaisha
- Inventors: Yusuke Hayashi; Taichi Kawanai; Daichi Hotta
- Web page: https://patentgrep.com/patent/US12112279B2

## CPC

- B60W 30/0956: {the prediction being responsive to traffic or environmental parameters}
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- B60W 30/08: {Active safety systems} predicting or avoiding probable or impending collision {or attempting to minimise its consequences}
- B60W 40/04: Traffic conditions
- G05B 13/0265: {the criterion being a learning criterion}
- B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)

## Abstract

A distribution calculation process, a similarity level calculation process, a similarity level evaluation process, and a collision determination process are executed. In the distribution calculation process, a probability density distribution related to a future position of a moving object is calculated for each behavior pattern using position data and speed data of the moving object, error data thereof, and a plurality of behavior models set for each behavior pattern corresponding to a type of the moving object. In the similarity level calculation process, a similarity level between the probability density distribution and a reference probability density distribution is calculated for each behavior pattern. In the similarity level evaluation process, the similarity level is evaluated. The similarity level includes an out-of-range level falling below a similarity allowable range. In the similarity level evaluation process, the behavior pattern having the out-of-range level is excluded from a target of the collision determination process.

## Claims

### Claim 1

A behavior prediction device that predicts a future behavior of a moving object around a vehicle, the moving object being detected by the vehicle, the behavior prediction device comprising:
a database that stores a plurality of behavior models set for each behavior pattern corresponding to a type of the moving object;
a memory that stores position data and speed data of the moving object and error data indicating an error of the position data and the speed data; and
a processor, wherein:
the processor is configured to execute:
a distribution calculation process of calculating a probability density distribution related to a future position of the moving object for each of the behavior patterns using the position data, the speed data, the error data, and the behavior models,
a similarity level calculation process of calculating a similarity level between the probability density distribution and a reference probability density distribution for each of the behavior patterns, the reference probability density distribution indicating the probability density distribution calculated using the behavior model corresponding to a current behavior of the moving object at a processing timing prior to a current processing timing of the distribution calculation process,
a similarity level evaluation process of evaluating the similarity level,
an error data changing process of changing the error data based on the similarity level,
a collision determination process of determining a collision between the vehicle and the moving object using the probability density distribution, and
upon determination of the collision between the vehicle and the moving object, actuating at least one of a braking actuator and a steering actuator to cause the vehicle to perform a collision avoidance operation;
the similarity level includes an out-of-range level that falls below a similarity allowable range;
the similarity level includes a low similarity level that is present within the similarity allowable range and falls below a low similarity threshold set within the similarity allowable range;
in the similarity level evaluation process, the processor excludes the behavior pattern having the out-of-range level from a target of the collision determination process;
in the error data changing process, the processor increases the error data to be used at a next processing timing of the distribution calculation process using a low similarity model as compared with the error data used at the current processing timing of the distribution calculation process; and
the low similarity model indicates the behavior model corresponding to the behavior pattern having the low similarity level.

### Claim 2 (depends on 1)

The behavior prediction device according to claim 1, wherein:
the similarity level includes a high similarity level that is present within the similarity allowable range and exceeds a high similarity threshold set within the similarity allowable range;
the processor is further configured to execute a distribution recalculation process of recalculating the probability density distribution using a high similarity model indicating the behavior model corresponding to the behavior pattern having the high similarity level;
in the distribution recalculation process, the processor executes a superposition of the probability density distribution calculated using the high similarity level at the current processing timing on the probability density distribution calculated using the high similarity level at a processing timing prior to the current processing timing; and
in the collision determination process, the processor determines the behavior pattern having the high similarity level based on the probability density distribution after the superposition by the distribution recalculation process.

### Claim 3

A behavior prediction device that predicts a future behavior of a moving object around a vehicle, the moving object being detected by the vehicle, the behavior prediction device comprising:
a database that stores a plurality of behavior models set for each behavior pattern corresponding to a type of the moving object;
a memory that stores position data and speed data of the moving object and error data indicating an error of the position data and the speed data; and
a processor, wherein:
the processor is configured to execute
a distribution calculation process of calculating a probability density distribution related to a future position of the moving object for each of the behavior patterns using the position data, the speed data, the error data, and the behavior models,
a similarity level calculation process of calculating a similarity level between the probability density distribution and a reference probability density distribution for each of the behavior patterns, the reference probability density distribution indicating the probability density distribution calculated using the behavior model corresponding to a current behavior of the moving object at a processing timing prior to a current processing timing of the distribution calculation process,
a similarity level evaluation process of evaluating the similarity level,
an error data changing process of changing the error data based on the similarity level,
a collision determination process of determining a collision between the vehicle and the moving object using the probability density distribution, and
upon determination of the collision between the vehicle and the moving object, actuating at least one of a braking actuator and a steering actuator to cause the vehicle to perform a collision avoidance operation;
the similarity level includes an out-of-range level that falls below a similarity allowable range;
the similarity level includes a high similarity level that is present within the similarity allowable range and exceeds a high similarity threshold set within the similarity allowable range;
in the similarity level evaluation process, the processor excludes the behavior pattern having the out-of-range level from a target of the collision determination process;
in the error data changing process, the processor decreases the error data to be used at a next processing timing of the distribution calculation process using a high similarity model as compared with the error data used at the current processing timing of the distribution calculation process; and
the high similarity model indicates the behavior model corresponding to the behavior pattern having the high similarity level.

### Claim 4

A behavior prediction device that predicts a future behavior of a moving object around a vehicle, the moving object being detected by the vehicle, the behavior prediction device comprising:
a database that stores a plurality of behavior models set for each behavior pattern corresponding to a type of the moving object;
a memory that stores position data and speed data of the moving object and error data indicating an error of the position data and the speed data; and
a processor, wherein:
the processor is configured to execute
a distribution calculation process of calculating a probability density distribution related to a future position of the moving object for each of the behavior patterns using the position data, the speed data, the error data, and the behavior models,
a similarity level calculation process of calculating a similarity level between the probability density distribution and a reference probability density distribution for each of the behavior patterns, the reference probability density distribution indicating the probability density distribution calculated using the behavior model corresponding to a current behavior of the moving object at a processing timing prior to a current processing timing of the distribution calculation process,
a similarity level evaluation process of evaluating the similarity level,
a collision determination process of determining a collision between the vehicle and the moving object using the probability density distribution, and
upon determination of the collision between the vehicle and the moving object, actuating at least one of a braking actuator and a steering actuator to cause the vehicle to perform a collision avoidance operation;
the similarity level includes an out-of-range level that falls below a similarity allowable range;
the similarity level includes a high similarity level that is present within the similarity allowable range and exceeds a high similarity threshold set within the similarity allowable range;
in the similarity level evaluation process, the processor excludes the behavior pattern having the out-of-range level from a target of the collision determination process;
the processor is further configured to execute a distribution recalculation process of recalculating the probability density distribution using a high similarity model indicating the behavior model corresponding to the behavior pattern having the high similarity level;
in the distribution recalculation process, the processor executes a superposition of the probability density distribution calculated using the high similarity level at the current processing timing on the probability density distribution calculated using the high similarity level at a processing timing prior to the current processing timing; and
in the collision determination process, the processor determines the behavior pattern having the high similarity level based on the probability density distribution after the superposition by the distribution recalculation process.